Gliomas are the most common primary neoplasm of the human central nervous system. They may be of astrocytic, oligodendroglial, or mixed origin. Different malignancy grades can be distinguished histopathologically, with the most malignant grade (glioblastoma multiforme, grade IV glioma) being uniformly fatal despite aggressive surgery, radiotherapy, and chemotherapy. Prognosis is somewhat
From the Division of Medical Genetics, Department of Medi- cine, McGill University and Montreal General Hospital Research Institute (D. W., S. R., G. A. R.), Centre for Research in Neuro- science, McGill University and the Montreal General Hospital, Montreal, Canada (M. H. R., J. S,, G. A. R.); and Clinique Neu- rologique, H6pital de la SallaetriOre, (M. P., J.- Y. D.) Paris, France.
Address reprint requests to: David Watkins, PhD, Division of Medical Genetics, Montreal General Hospital, 1650 Cedar Ave- nue, Room L10-107, Montreal, Quebec, Canada H3G 1A4.
Received December 3, 1995; accepted May 13, 1996.
Cancer Genet Cytogenet 92:73-78 (1996) O Elsevier Science Inc,, 1996 655 Avenue of the Americas, New York, NY 10010
better for less malignant gliomas, but there is a tendency for relatively benign tumors to progress over time to a more malignant phenotype.
The molecular genetic alterations underlying genesis and progression of human gliomas are beginning to be elu- cidated. Both activation of cellular proto-oncogenes and inactivation of tumor suppressor genes have been impli- cated. Loss of heterozygosity (LOH) suggestive of inactiva- tion of tumor suppressor genes has been observed with particular frequency for loci on chromosomes lp, 9p, 10q, 13q, 17p, 19q and 22q [1-9]. Inactivation of the TP53 gene on 17p, which encodes the p53 tumor suppressor, is clearly implicated in sporadic and familial forms of glioma [10, 11]. The region of LOH on 9p includes the MTS1 locus encoding p16, a known tumor suppressor, and the gene encoding a second potential tumor suppres- sor, p15. Deletion of both alleles of these genes occurs fre- quently, although point mutations of either gene are rare in gliomas [12, 13]. Tumor suppressor genes associated
0165-4608/96/$15.00 PII S0165-4608(96)00149-5
74 D. Watkins et al.
with glioma development have not yet been identified in the other regions of LOH.
There is cytogenetic and molecular evidence for loss of all or part of chromosome 22 in human gliomas [3, 6, 14- 17], but a tumor suppressor gene on chromosome 22 play- ing a role in development of human gliomas has not been identified. The neurofibromatosis type 2 (NF2) tumor sup- pressor gene has been localized to 22q and cloned [18, 19]. Germ line mutations of this gene result in increased sus- ceptibility to development of several types of neurological tumors, including schwannomas, meningiomas and, to a lesser extent, gliomas [20]. This gene thus represents an excellent candidate for a tumor suppressor gene inacti- vated during development of sporadic gliomas. Additional tumor suppressor genes may also be present on 22q and im- plicated in development of gliomas. Terminal deletions of 22q distal to the NF2 locus have been identified in a small proportion of malignant gliomas and may signal the exist- ence of a tumor suppressor gene here [6]. Evidence for an- other tumor suppressor gene on 22q that is inactivated in sporadic meningiomas remains controversial [21-23]. At present, the NF2 gene remains the only well-characterized tumor suppressor gene known on this chromosome.
In the present report, we describe analysis of a panel of sporadic human gliomas for LOH on 22q, with the aim of better characterizing the segment (or segments) of this chromosome most often affected. We also describe direct investigation of the NF2 gene, using Southern blot and single strand conformation analysis (SSCA), to determine whether this gene is frequently inactivated during devel- opment of gliomas.
MATERIALS AND METHODS
Tumor Acquisition and Preparation of DNA Tumor samples were obtained from 70 patients undergoing surgery for resection of glioma. Samples were immediately placed in liquid nitrogen and stored at - 80 ° until DNA extraction was performed. In most cases, a blood sample was obtained from the patient at the same time as the tumor sam- ple. Most tumors were of astrocytic origin, but there were 5 oligodendrogliomas. Tumors were classified according to a D22S1 four-tier system with grade IV gliomas (glioblastoma mul- D22S9 tiforme) being the most malignant. There were four grade D22S15 I, 13 grade II, 13 grade III, and 37 grade IV tumors in the D22S21 panel; there were also two giant cell astrocytomas from D22S22 patients with tuberous sclerosis and one unclassified D22S23 glioma in the panel. DNA was isolated from tumor and D22S24 blood samples using previously described methods [24]. D22S28
D22S29 Southern Blotting D22S32
D22S45 The loci tested for LOH and the restriction endonucleases D22S80 used to detect sequence polymorphisms at these loci are D22S94 listed in Table 1. The order of these markers on the long BCR arm of chromosome 22 was determined on the basis of CRYB2 published reports of linkage and physical mapping studies GGT [18, 26, 37-39]. An unambiguous order could be estab- MB lished for most markers, with two exceptions. GGT has NEFH been localized to the proximal part of 22q distal to BCR NF2 [39], but its localization relative to other markers in the PDGFB
region is not known; and the relative order of PDGFB and D22S80 could not be determined. Inferred order of mark- ers on chromosome 22 is shown in Figure 1. These markers cover the entire long arm of chromosome 22 from proximal q2211.2 (D22S24 and D22S9) to 22q13.3 (D22S94, D22S22, D22S23, D22S45, and D22S21).
Blood and tumor DNA was digested with the appropriate restriction endonuclease using reaction conditions specified by the manufacturer. The resulting DNA fragments were electrophoresed into 0.8% agarose gels in Tris/borate/EDTA (TBE) electrophoresis buffer. DNA was denatured and trans- ferred to nylon membranes, following established protocols. Probes detecting sequence polymorphisms were labeled by random oligonucleotide priming in the presence of al- pha[32p]dATP and hybridized to filters, under stringent conditions. Filters were then washed and exposed to x-ray film at -80 °. Constitutional (blood) and tumor genotypes for the same individual were compared to determine whether LOH had occurred during tumor development.
To determine whether large scale rearrangements of the NF2 gene had occurred during tumorigenesis, tumor DNA digested with EcoRI or HincII was electrophoresed and blotted onto nylon membranes as above, and hybridized with radioactively labeled NF2 cDNA probe N1.1 [18].
Single Strand Conformation Analysis SSCA was performed on genomic DNA from tumors. Each exon of the NF2 gene was amplified by the polymerase chain reaction (PCR) in the presence of alpha[35S]dATP, using primers recognizing flanking intronic sequences. Primer sequences have been previously published [40, 41]. Labeled PCR product was denatured and run on 6% non- denaturing polyacrylamide gels (29:1 acrylamide:bisacryl- amide) at 4 °. Each sample was analyzed twice, once in a gel containing 10% glycerol and once in a gel without gly-
Table I Chromosome 22 DNA polymorphisms used in the study
Glil Gli3 Gli4 Gli5 Gi l l5 Gli19 Gli21 Gli22 Gli26 Gli29 Gli32 Gli41 Gli59 Gli64 Gli71
D22S24
D22S9 BCR
GGT
CRYB2
D22S1
NEFH
NF2 D22S32
D22S15 D22S28
D22S29
MB
D22S80 D22S94
D22S22 D22S23 D22S45 D22S21
i i i I I
Figure I Extent of loss of heterozygosity on chromosome 22 in 15 sporadic human gliomas. Markers are shown in order from centromere (top) to telomere (bottom) of the long arm of the chromosome. Solid bars indicate regions of chromosome 22 for which LOH was definitely shown; hatched bars indicate regions for which markers were uninformative, in which LOH may or may not have occurred. A: region to which the NF2 gene is localized; B: most frequently shared region of LOH.
A
B
carol [42]. Gels were dr ied and exposed to x-ray film at room temperature for 72-120 hours.
RESULTS
Loss of Heterozygosity StEudies A panel of 47 b lood- tumor pairs (one grade t, 10 grade II, eight grade III and 27 grade IV tumors) was tested for LOH at 22 loci on chromosome 22. Al l gliomas in the panel were informative for at least some of these markers. LOH was observed in 15 gliomas, inc luding 0 of two grade I, two of 10 grade II, three of eight grade III and 10 of 27 grade IV tumors. Al l were of astrocytic origin. In four tumors (G]i 1, Gli 5, Gli 32, and Gli 64), LOH was observed for all informative markers (Figure 1), suggesting that one entire copy of chromosome 22 had been lost during tumor- igenesis. In the remaining tumors, smaller segments of the chromosome were affected. No common area of LOH shared by all 15 tumors could be identif ied. The common area of LOH shared by the largest number of gliomas in the
panel was bounded proximal ly by D22S15 and dis tal ly by MB. This region was defini tely affected in nine tumors. In two addi t ional tumors, markers in this region were unin- formative, but there was LOH for markers adjacent to the region; LOH may have occurred in the region of interest as well. The tumors affected inc luded one grade II, two grade III, and eight grade IV tumors.
LOH at the NF2 locus was tested direct ly using the probe N1.1, which detects a three-al lele MspI polymor- ph i sm (unpubl i shed data). In eight gl iomas there was LOH of the NF2 locus; in two addi t ional tumors wi th LOH at adjacent loci, the N1.1 probe was uninformative, and loss of one copy of the NF2 gene could not be def ini t ively es- tabl ished or rejected (Figure 1). Gliomas wi th definite or poss ible LOH at the NF2 locus inc luded one grade II, three grade III, and six grade IV tumors.
Single Strand Conformation Analysis SSCA was undertaken, using genomic DNA from the entire panel of 70 gliomas. No change in migrat ion pat tern
76 D. Watkins et al.
suggestive of a mutat ion in the NF2 gene was detected in any of the tumors of the panel.
DISCUSSION
Results of cytogenetic and molecular genetic studies have suggested that inactivation of tumor suppressor genes plays an important role in the genesis and progression of gliomas in humans. Inactivation of the TP53 tumor sup- pressor gene on 17p clearly occurs in the development of many gliomas, and evidence suggests that deletion of the genes encoding p16 and p15 on 9p may also be playing a role. In the present study, the frequency and localization of LOH on 22q was analyzed in a panel of human gliomas, and the NF2 gene was tested for potentially inactivating mutations.
LOH was observed in 15 of 47 gliomas in our panel, in- cluding tumors of grades II, III, and IV. Frequency of LOH was not significantly different between grade II, III, and IV gliomas (X2 analysis); because only two grade I tumors were available, it was not possible to assess the rate of LOH in tumors of this grade. These results suggest that LOH on 22q is associated with an early stage of glioma de- velopment rather than with progression of gliomas to a highly malignant state. This contrasts with LOH for chro- mosome 10 markers, which in our study and others has been observed at high frequency in high grade (grade III and IV) gliomas, but never in low grade tumors [1, 2, 4, 7].
No single commo.n region of LOH could be identified in our panel of gliomas. LOH affecting the NF2 locus was present in at least eight, and possibly 10, tumors. The most frequent affected segment, bounded proximally by D22S15 and distally by MB, was lost in at least nine, and possibly 11, tumors in our panel and lies distal to the NF2 locus. The lack of any single common region of LOH among our tumors with chromosome 22 LOH suggests that there may be more than one tumor suppressor gene on this chromosome that can play a role in development of glio- mas, or alternatively that at least some cases of LOH on chromosome 22 were not associated with inactivation of a tumor suppressor gene.
Since LOH for the NF2 locus was observed in as many as 21% of the tumors in our panel, and since germ line mutat ions of the NF2 gene are associated with an in- creased tendency to develop gliomas, we investigated the NF2 gene in an extended panel of 70 gliomas. No gross re- arrangements of the gene were detected on Southern anal- ysis of restriction endonuclease-digested tumor DNA, and no abnormally migrating bands were detected by SSCA of genomic DNA in any of the tumors. Several mutat ions of the NF2 gene have been detected by SSCA in sporadic schwannomas and meningiomas, as well as in constitu- tional and tumor DNA from NF2 patients [40, 41, 43, 44]. Thus it appears that mutat ion of the NF2 gene is not a common occurrence in human gliomas, a result consistent with that of a previous study [45]. In no case was inactiva- tion of both copies of the NF2 gene detected. These results strongly suggest that inactivation of the NF2 gene is not a common event in development of sporadic gliomas in hu- mans, although it is theoretically possible that this type of
tumor is associated specifically with mutat ions that affect noncoding regions of the NF2 gene.
Terminal deletions of chromosome 22 have been ob- served by cytogenetic and molecular genetic means in a small proportion of high-grade gliomas [6], and it has been suggested that a tumor suppressor gene on 22q distal to D22S80 may play a role in tumor progression. LOH specif- ically affecting this portion of chromosome 22 was ob- served in only two of 47 tumors in our panel; this portion was also lost in six gliomas in which LOH affected all or nearly all of the chromosome. These results do not strongly support the existence of a tumor suppressor gene distal to D22S80 associated with glioma progression.
These results confirm that LOH for loci on chromosome 22q occurs with some frequency in huma n gliomas of grades II, III, and IV, although the NF2 gene is apparently excluded as the affected tumor suppressor gene. The seg- ment of 22q bounded by D22S15 and MB, distal to NF2, represents a possible area in which a tumor suppressor gene may be localized. No likely candidate genes have, as yet, been identified in this area.
The authors thank Drs. J.-G. Villemure, P. Muller, R. Ford, J.-L. Caron, P. Richardson, and R. Del Maestro for providing brain tumor and normal brain samples used in this study. D.W. is sup- ported by a scholarship from the Montreal Children's Hospital Research Institute.
REFERENCES
1. James CD, Carlbom E, Dumanski JP, Hansen M, NordenskjSld M, Collins VP, Cavenee WK (1988): Clonal genomic alter- ations in glioma malignancy stages. Cancer Res 48:5546- 5551.
2. El-Azouzy M, Chung RY, Farmer GE, Martuza RL, Black RM, Rouleau GA, Hettlich C, Hedley-White ET, Zervas NT, Pana- gopoulos K, Nakamura Y, Gusella JF, Seizinger B (1989): Loss of distinct regions on the short arm of chromosome 17 associ- ated with tumorigenesis of human astrocytomas. Proc Natl Acad Sci USA 86:7186-7190.
3. Fults D, Pedone CA, Thomas GA, White R (1990): Allelotype of human malignant astrocytoma. Cancer Res 50:5784-5789.
4. Fults D, Brockmeyer D, Tallous MW, Pedone CA, Cawthon RM (1992): p53 mutation and loss of heterozygosity on chro- mosomes 17 and 10 during human astrocytoma progression. Cancer Res 52:674-679.
5. Olopade OI, Jenkins RB, Ransom DT, Malik K, Pomykala H, Nobori T, Cowan JM, Rowley JD, Diaz M (1992): Molecular analysis of deletions of the short arm of chromosome 9 in human gliomas. Cancer Res 52:2523-2529.
6. Rey JA, Bello MJ, Jimenez-Lara AM, Vaquero J, Kusak ME, de Campos JM, Sarasa JL, Pestafia A (1992): Loss of heterozygos- ity for distal markers on 22q in human gliomas. Int J Cancer 51:703-706.
7. Rasheed BKA, McLendon RE, Herndon JE, Friedman HS, Friedman AH, Bigner DD, Bigner SH (1994): Alterations of the TP53 gene in human gliomas. Cancer Res. 54:1324-1330.
8. Rubio M-P, Correa KM, Ueki K, Mohrenweiser HW, Gusella JF, yon Deimling A, Louis DN (1994): The putative tumor suppressor gene on chromosome 19q maps between APOC2 and HRC. Cancer Res 54:4760-4763.
9. Louis DN, Gusella JF (1995): A tiger behind many doors: mul- tiple genetic pathways to malignant glioma. Trends Genet 11:412-415.
The NF2 Gene in H u m a n Gl iomas 77
10. Franke] RH, Bayona W, Koslow M, Newcomb EW (1992): p53 mutations in human malignant gliomas: comparison of loss of heterozygosity with mutation frequency. Cancer Res 52:1427-1433.
11. Malkin D, Li FP, Strong LC, Fraumeni JF, Nelson CE, Kim DH, Kassel J, Gryka MA, Bischoff FZ, Tainski MA, Friend SH (1990): Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas and other neoplasms. Science 250:1233-1238.
12. Ueki K, Rubio M-P, Ramesh V, Correa KM, Rutter JL, von Deimling A, Buckler AJ, Gusella JF, Louis DN (1994): MTS1/ CDKN2 gene mutations are rare in primary human astrocyto- mas with allelic loss of chromosome 9p. Hum. Molec Genet 3:1841-1845.
13. Jen J, Harper JW, Bigner SH, Bigner DD, Papadopoulos N, Markowitz S, Willson J]~V, Kinzler KW, Vogelstein B (1994): Deletion of p16 and p15 genes in brain tumors. Cancer Res 54:6353-6358.
14. Rey JA, Bellow MJ, de Campos JM, Kusak ME, Ramos C, Ben- itez J (1987): Chromosomal patterns in malignant human astrocytomas. Cancer Genet Cytogenet 29:201-221.
15. Rey JA, Bello MJ, de Campos JM, Kusak ME, Moreno S (1987): Chromosomal compositions of a series of 22 low grade gliomas. Cancer Genet Cytogenet 29:223-237.
16. Bigner SH, Mark J, BcLrger PC, Mahaley MS, Bullard DE, Muhlbaier LH, Bigner DD (1988): Specific chromosomal abnormalities in malignant human gliomas. Cancer Res 48:405-411.
17. Jenkins RB, Kimmel DW, Moertal CA, Schultz CG, Scheu- rauer BW, Kelly PJ, Deald GW (1989): A cytogenetic study of 53 human gliomas. Cancer Genet Cytogenet 39:253-279.
18. Rouleau GA, Merel P, Lutchman M, Sanson M, Zucman J, Marineau C, Hoang-Xuan K, Demczuk S, Desmaze C, Plou- gastel B, Pulst S, Lenoir G, Bijlsma E, Fashold R, Dumanski J, de Jong P, Parry D, Eldridge R, Aurias A, Delattre O, Thomas G (1993): Alteration in aL new gene encoding a putative mem- brane-organizing protein causes neurofibromatosis type 2. Nature 363:515-521.
19. Trofatter JA, MacCollir~ MM, Rutter JL, Murrell JR, Duyao MP, Parry DM, Eldridge R, Kley N, Menon AG, Pulaski K, Haase VH, Ambrose CM, Munroe D, Bove C, Haines JL, Mar- tuza RL, MacDonald ME, Seizinger BR, Short MP, Buckler AJ, Gusella JF (1993): A novel moesin-, ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor sup- pressor. Celt 72:791-800.
20. Evans DGR, Hnson SM, Donnai D, Neary W, Blair V, Newton V, Harris R (1992): A clinical study of type 2 neurofibromato- sis in the United Kingdom. I. Prevalence, mutation rate, fit- ness and confirmation of maternal transmission effect on severity. J Med Genet 2cJ:841-846.
21. Dumanski JP, Ronleau GA, Nordenskj61d M, Collins VP (1990): Molecular genetic analysis of chromosome 22 in 81 cases of meningioma. Cancer Res 50:5863-5867.
22. Lekanne Deprez RH, Groen NA, van Biezen NA, Hagemeijer A, van Drunen E, Koper JW, Avezaat CJJ, Bootsma D, Zwar- thoff EC (1991): A t(4;22) in a meningioma points to the localization of a putative tumor suppressor gene. Am J Hum Genet 48: 783-790.
23. Sanson M, Marineau C, Desmaze C, Lutchman M, Ruttledge M, Baron C, Narod S, De,lattre O, Lenoir G, Thomas G, Aurias A, Rouleau GA (1993): Germline deletion in a neurofibroma- tosis type 2 kindred inactivates the NF2 gene and a candidate meningioma locus. Hum Molec Genet 2:1215-1220.
24. Seizinger BR, Rouleau GA, Ozelius LJ, Lane AH, St George- Hyslop P, Huson S, Gusella JF, Martuza RL (1987): Common
pathogenic mechanism for three tumor types in bilateral acoustic neurofibromatosis. Science 236:317-319,
25. Rouleau GA, Kurnit DM, Neve RL, Bazanowski A, Patterson D, Gusella JF (1988): D22S15--a fetal brain cDNA with BanI and SacI RFLP. Nucl Acids Res 16:1646.
26. Rouleau GA, Haines JL, Bazanowski A, Colella-Crowley A, Trofatter JA, Wexler NS, Conneally PM, Gusella JF (1989): A genetic linkage map of the long arm of human chromosome 22. Genomics 4:1-6.
27. Krapcho K, Nakamura Y, Fujimoto E, Eldridge R, Leppert M, O'Connell P, Lathrop GM, Lalouel J-M, White R (1988): Isola- tion and mapping of a polymorphic DNA sequence (pEFZ31) on chromosome 22 (D22S32). Nucl Acids Res 16:5704.
28. Budarf M, McDermid H, Sellinger B, Emmanuel BS (1988): Isolation, RFLP analysis and regional mapping of anonymous probes to human chromosome 22. Am J Hum Genet 43:A139.
29. Zhang F, Aurias A, Delattre O, Thomas G (1989): RFLP iden- tified by the anonymous DNA segment FZ V D l l at 22q13 (HGM no, D22S80). Nucl Acids Res 17:1791.
30. Dumanski JP, Guerts von Kessel AHM, Ruttledge M, Wladis A, Sngawa N, Collins VP, NordenskjSld M (1990): Isolation of anonymous polymorphic DNA fragments from human chromosome 22q12-qter. Hum Genet 84:219-222.
31. Roschmann E, Assure G, Fink T (1987): RFLP detected with a 5'-bcr-gene-sequence [HGM provisional no. D22Sll]. Nucl Acids Res 15:1883.
32. Fontaine B, Harvey M, Haines JL, Gorin MB, Gusella JF, Rou- leau GA (1989): A human sequence which cross hybridizes to bovine B2 crystallin displays an informative PstI polymorphism and maps to chromosome 22. Cytogenet Cell Genet 51:999.
33. Rouleau GA, Bazanowski A, Cohen EH, Guellaen G, Gusella JF (1988): Gamma glutamyl transferase locus (GGT) displays a PvulI polymorphism. Nucl Acids Res 16:11848.
34. Julier C, Lathrop GM, Lalouel J-M, Kaplan JC (1985): Use of multi-locus tests of gene order: example for chromosome 22. Cytogenet Cell Genet 40:663-664.
35. Lacoste-Royal G, Mathieu M, Ganvreau D (1989): HincII RFLP in the human gene for the heavy neurofilament subunit (NF-H). Nucl Acids Res 17:6434.
36. Fourney RM, Dietrich KD, Aubin RA, Paterson MC (1988): HindIII polymorphism in the human c-sis proto-oncogene. Nucl Acids Res 16:8197.
37. Delattre O, Azambuja CJ, Aurias A, Zucman J, Peter M, Zhang F, Hors-Cayla MC, Rouleau G, Thomas G (1991): Mapping of human chromosome 22 with a panel of somatic cell hybrids. Genomics 9:721-727.
38. Dumanski JP, Carlbom E, Collins VP, NordenskjSld M, Emmanuel BS, Budarf ML, McDermid HE, Wolff R, O'Con- nell P, White R, Lalouel J-M, Leppert M (1991): A map of 22 loci on human chromosome 22. Genomics 11:709-719.
39. Collins JE, Everett LA, Bentley DR, Dunham I (1992): A panel of human chromosome 22 specific sequence tagged sites. Genomics 14:1098-1103.
40. Ruttledge MH, Sarrazin J, Rangaratnam S, Phelan CM, Twist E, Merel P, Delattre O, Thomas G, Nordenskj61d M, Collins VP, Dumanski JP, Rouleau GA (1994): Evidence for the com- plete inactivation of the NF2 gene in a majority of sporadic meningiomas. Nature Genet 6:180-184.
41. Jacoby LB, MacCollin M, Louis DN, Mohney T, Rubio M-P, Pulaski K, Trofatter JA, Kley N, Seizinger B, Ramesh V, Gusella JF (1994): Exon scanning for mutation of the NF2 gene in schwannomas. Hum Molec Genet 3:413-419.
78 D. Watk ins et al.
42. Triggs-Raine BL, Akerman BR, Clarke JTR, Gravel RA (1991): Sequence of DNA flanking the exons of the HEXA gene, and identification of mutations in Tay-Sachs disease. Am J Hum Genet 49:1041-1054.
43. Twist EC, Ruttledge MH, Rousseau M, Papi L, Sanson M, Merel P, Delattre O, Thomas G, Rouleau GA (1994): The neu- rofibromatosis type 2 gene is inactivated in schwannomas. Hum Molec Genet 3:147-151.
44. MacCollin MM, Ramesh V, Jacoby LB, Louis DN, Rubio M-P, Pulaski K, Troffater JA, Short MP, Bove C, Eldridge R, Parry DM, Gusella JF (1994): Mutational analysis in patients with neurofibromatosis 2. Am J Hum Genet 55:314-320.
45. Rubio M-P, Correa KM, Ramesh V, MacCollin MM, Jacoby LB, von Deimling A, Gusella JF, Louis DN (1994b): Analysis of the neurofibromatosis 2 gene in human ependymomas and astrocytomas. Cancer Res 54:45-47.
